Does the guilty plea system produce results different from those that would occur at trial? To be more specific, does it convict defendants who are in fact innocent (and would be acquitted) or convict defendants who committed the offense but could not be found guilty beyond a reasonable doubt at trial? Does the plea negotiation process produce convictions for offenses significantly different from the crimes a defendant may actually have committed? For the many observers who are inclined to give affirmative answers to these questions, the entire body of substantive criminal law may seem a supreme irrelevance; convictions obtained by guilty pleas are seen as the outcome of hurried horse-trading rather than the thoughtful application of complex legal principles to the known facts. What is the practical significance of substantive law in a world dominated by plea bargaining? Consider the following comments:

A General View of the Criminal Justice System This chart seeks to present a simple yet comprehensive view of the movement of cases through the criminal justice system. Procedures in individual jurisdictions may vary from the pattern shown here. The differing weights of line indicate the relative volumes of cases disposed of at various points in the system, but this is only suggestive since no nationwide data of this sort exists.

p. 19

1. Conflicts of interest. What sorts of “lawyers’ self-interest” could skew the bargaining process in the ways that Professor Bibas describes? Can you think of personal motivations that might (consciously or subconsciously) lead a defense attorney to advise her client to accept an unduly severe plea offer? Can you think of personal motivations that might lead a prosecutor to offer an overly lenient plea offer? Would it ever be in a defense attorney’s self-interest to recommend rejection of a favorable plea agreement?

2. Agency costs. Where potential conflicts of interest exist, monitoring is essential; there must be some way to assure that attorneys act in the best interests of their clients—the accused in the case of the defense attorney and the general public in the case of the prosecutor. What factors make it difficult for some (or all) defendants to determine whether their attorney’s advice is influenced by the attorney’s personal interests? What factors make it difficult for the general public to determine whether the prosecutor’s actions are influenced by her personal interests? For further discussion of the p. 23problems of attorney self-interest and agency costs, see Stephen J. Schulhofer, Criminal Justice Discretion as a Regulatory System, 17 J. Legal Stud. 43 (1988).

3. Implications for substantive criminal law. Despite the distortions described by Alschuler and Bibas, there undoubtedly remains a degree of truth in the picture painted by Enker: At least to some extent, the substantive law and the expected outcome of a trial remain important elements (though not the only elements) in determining the outcome of plea negotiation. At this early point in the study of criminal law, it is premature to discuss plea bargaining in greater detail or to investigate possibilities for reform. We defer those issues to Chapter 10, where we consider the ways in which the modern administrative state uses prosecutorial discretion, plea bargaining, and sentencing practices to determine culpability and impose punishment outside the framework of the formal criminal trial.

For present purposes, our concern is to set the stage for the study of substantive criminal law by asking whether its principles really make a difference in a world dominated by plea bargaining. On that question, there can be no doubt that criminal law matters, but it is not the only thing that matters. And we can expect the role of the substantive law to grow when cases are handled by sophisticated attorneys who know the doctrines of criminal law well enough to invoke them effectively in the course of bargaining.

Sections B and C of this chapter described the organization of the criminal justice system and typical pretrial procedures. To introduce our examination of the trial stage itself, the present note briefly summarizes the typical course of a formal criminal trial. Naturally, the procedure followed in some jurisdictions or in particular cases may differ in points of detail from that set out in this preliminary overview.

A formal trial typically begins with the selection of the jury. A panel of prospective jurors (called a venire) enters the courtroom, and the judge describes the nature of the case and the identity of the parties so that any prospective jurors who are personally involved may be excused. Prospective jurors are then questioned individually by the judge or by opposing counsel to determine possible bias. On the basis of this questioning (called voir dire) prospective jurors may be excused for cause, and both prosecution and defense may remove a certain number of the panel, without showing cause, by exercising peremptory challenges. When the requisite number of acceptable jurors (usually 12) has been obtained by this procedure, the panel is sworn.

Now the presentation of the case begins. Usually, the indictment is read to the jury; the prosecutor then makes an opening statement outlining the facts she plans to prove. Defense counsel also may make an opening statement. Claims made in these statements do not constitute evidence; they serve only to help the jury understand the testimony about to be presented. In many p. 24jurisdictions, motions relating to the scope or validity of the indictment and motions to suppress evidence must be made prior to trial. In others, such motions can be made at the outset of the trial. If the judge grants a motion to dismiss the indictment, the case will terminate at this point. Otherwise the trial goes forward, although motions may be made or renewed as the trial proceeds.

Next the prosecution calls its witnesses. Their testimony often evokes objections from counsel, and the judge must decide the bounds of permissible testimony under complex rules of evidence. We examine some of these rules in the sections to follow. When the prosecution has completed the presentation of its evidence, the defense may choose to stand on the presumption of innocence and move for a directed verdict or judgment of acquittal on the ground that the charges have not been proved beyond a reasonable doubt. If such a motion is not made, or if the motion is made and denied, the defense may decide to offer its own evidence (through the defendant or other witnesses). The prosecution then will have an opportunity to present further evidence in rebuttal. When both sides have finished presenting their evidence, counsel may make closing arguments to the jury. Ordinarily, the prosecution’s closing argument is presented first (because it bears the burden of proof), and the prosecutor is usually allowed an opportunity for rebuttal after the closing argument of the defense.

At this point the work of the opposing parties has been completed. The judge now intercedes with her most important contribution—the instructions to the jury. Ordinarily, judicial comment on the evidence tends to be cautious and limited. Formal instructions on the law, however, are always given, and they are typically quite detailed. The instructions serve both as a guide for the jury’s deliberations and as a focal point for challenges on appeal in the event of a conviction. Usually, the instructions cover responsibilities of the jurors (for example, how to elect a foreperson, when to refrain from discussing the case with other jurors or outsiders), matters related to the relevance of particular kinds of testimony, and, above all, detailed explanations of the substantive criminal law applicable to the case, including the facts necessary to establish the offense and definitions of legal concepts that the jury is called on to apply.

The jury at last retires to deliberate. Its verdict of guilty or not guilty on each charge must be reached by a substantial majority (and usually by unanimity). After a verdict of guilty and the imposition of sentence (usually by the judge), the trial terminates. There may of course be an appeal. If it finds trial errors, the reviewing court may reverse the conviction and order another trial, since American double jeopardy principles generally do not bar the retrial of a defendant who has successfully appealed her conviction.45 A verdict of not guilty is not subject to review of any kind, even when flagrant errors p. 25prejudicial to the prosecution occurred at trial.46 When the jury is unable to agree by the requisite majority on a verdict of either guilty or not guilty (a hung jury case), a mistrial is declared, and the defendant then may be retried at the prosecutor’s discretion.

Such, in rough outline, is the procedure by which guilt must be established when a criminal case is fully litigated through the formal trial stage. Further treatment of trial procedure lies beyond the province of this book. Nevertheless, a book devoted primarily to the substantive law should provide an introduction to the central features of the process of proof, in order to illuminate the context in which criminal law is applied and the ways in which the procedural system shapes the substantive content of the criminal law. The remaining sections of this chapter undertake such an exploration.

1. The order of proof. At trial the evidence is presented in a formally prescribed order. The prosecution first calls witnesses in an effort to prove the elements of the offense charged. As we have seen, p. 24 supra, the case may terminate at this point if the prosecution fails to sustain its burden of proof. If not, the defense may then call witnesses to refute the prosecution’s case-in-chief or to establish some affirmative defense. The prosecution has an opportunity to recall witnesses or to call new witnesses for purposes of rebuttal, that is, to refute evidence offered by the defense. The defense in turn is afforded a chance to answer by rejoinder any matters introduced in the prosecutor’s rebuttal.

Within these stages the examination of each witness follows a similar pattern. The witness is first questioned by the party that called the witness (direct examination), and afterward the opposing party has an opportunity to question that witness (cross-examination). Further questioning by the first party (re-direct) and by the opposing side (re-cross) may follow.

2. Relevance. The rules governing the admissibility of evidence are extremely complex. For present purposes, it will be useful to begin by focusing on one obvious but deceptively simple requirement, the rule of relevancy. Irrelevant evidence is never admissible. The converse cannot be stated so categorically. Relevant evidence is generally admissible, but there are many exceptions to this principle. Before turning to the exceptions, we must first be clear about the meaning of relevancy.

p. 26Evidence is considered relevant for purposes of the rules of evidence only if it is both probative and material, and these are precise terms of art. Evidence is probative only if it tends to establish the proposition for which it is offered or—to be precise—if the proposition is more likely to be true given the evidence than it would be without the evidence. Thus, if the proposition to be proved is that H was the person who killed his wife W, evidence of a motive (that is, that H stood to inherit a substantial estate on W’s death) is probative. Of course, the existence of the motive does not, by itself, make it probable that H is the killer, but H is more likely to be the killer if he had a motive than if he did not; this greater likelihood is all that is required to establish probative value.

Probative value alone is often thought of as synonymous with relevancy. But relevancy for purposes of the rules of evidence requires in addition that the proposition that the evidence tends to prove be one that will affect the outcome of the case under applicable law. So, for example, evidence in a homicide prosecution that the defendant acted in self-defense is material, because under the substantive law self-defense is a defense. But evidence that the deceased consented to be killed is not material, because under the substantive law consent by the victim is not a defense to a homicide charge. Thus, evidence may be excluded as irrelevant for one of two distinct reasons—either because the evidence does not tend to establish the proposition in question or because that proposition is not material to the outcome of the case. The materiality requirement means that the first prerequisite for determining the relevancy and hence the admissibility of evidence is a command of the substantive law of crimes.

We may sum up what we have so far said about relevancy by quoting the formulation used in the Federal Rules of Evidence:

3. Privilege. When is relevant evidence not admissible? The law of evidence embodies dozens of distinct rules requiring the exclusion of relevant evidence. For example, the various rules relating to privilege give individuals the right to withhold certain kinds of testimony, often in order to protect particular interests of a witness or specially important relationships with others.

One of the most important privileges is the privilege against self-incrimination. The Fifth Amendment provides: “[N]or shall [any person] be compelled in any criminal case to be a witness against himself.” The Supreme Court has construed this provision to imply that the government cannot p. 27require a criminal defendant to take the witness stand, cannot invite a jury to draw adverse inferences from a defendant’s refusal to testify, and cannot in any other way compel the defendant to disclose potentially incriminating facts about the case. These principles are of great significance for the substantive criminal law. They not only make the prosecutor’s task more difficult than it otherwise would be, but they pose an especially difficult barrier when the defendant’s own frame of mind is an essential part of what the prosecutor must prove. The privilege against self-incrimination places largely beyond the government’s reach the best (and sometimes the only) information about these elusive state-of-mind facts. As a result, substantive criminal law must constantly consider whether the government should be obliged to prove state-of-mind facts, and if so, what kinds of evidence will satisfy that obligation.

4. Other exclusionary rules: prejudice. Like the rules of privilege, many other rules exclude relevant evidence. One of the best known is the rule barring the use of evidence obtained through an illegal search. Indeed, criminal lawyers often refer to this prohibition as the exclusionary rule, and it is a major topic of study in criminal procedure. We will consider some of the other exclusionary rules in connection with the substantive crimes for which they have the greatest significance.47 The balance of the present section is devoted to exploring one open-ended rule—the rule that evidence must be excluded whenever its probative value is outweighed by its prejudicial effect.

The term prejudicial effect has a technical meaning. To pursue the example previously mentioned, suppose that in H’s murder prosecution, testimony is offered that shortly before the discovery of W’s bullet-ridden body, H was seen running from the scene carrying a smoking revolver. This testimony will undoubtedly be harmful to H’s chances for acquittal, but it will not be prejudicial in the technical sense, because its harmfulness flows solely from its legitimate probative value. Evidence is considered prejudicial only when it is likely to affect the result in some improper way. Thus, prejudice is involved if the jury is likely to overestimate the probative value of the evidence or if the evidence will arouse undue hostility toward one of the parties.

The exclusionary rule relating to prejudicial effect is of pervasive importance in the substantive criminal law, and we will encounter it repeatedly in the chapters that follow. The principle of weighing probative value against prejudicial effect is particularly central to developing doctrines in such areas as rape-shield laws; the adequacy of victim provocation in homicide; battered spouse evidence; and testimony concerning intoxication, diminished capacity, and the insanity defense. As a foundation for the study of such topics throughout this book, the present section considers the problem of prejudicial effect in depth, in a context of recurring importance: evidence of other crimes.

p. 28The next case illustrates the principle of prejudicial effect and its relationship to basic conceptions of criminal responsibility.

1. On the facts of Zackowitz, we may assume that the killing was intentional. The principal issue at trial was whether the killing was “deliberate” (that is, whether the intent to kill was formulated before the shot, making the crime first-degree murder) or whether the killing was instead “impulsive” (that is, whether the intent to kill was formulated during the final scuffle, making the crime second-degree murder). Does the defendant’s possession of the weapons have some bearing on this issue? Recall that very slight probative value is usually sufficient to render evidence admissible; for example, evidence that Zackowitz stood to inherit money from the victim would undoubtedly be admissible as tending to show a motive for a deliberate killing. Does the evidence of weapons possession have at least that much probative value?

2. p. 31If the weapons evidence was relevant, as Judge (later Justice) Cardozo seems to assume, are there convincing reasons for excluding it? Judge Cardozo stresses that the issue in the case was a “delicate” and “subtle” one. Under these circumstances was it not particularly important for the jury to have access to as much relevant evidence as possible? In which case will the trier of fact be better able to evaluate what actually happened: when it knows about the weapons, knows the defendant’s explanation for them and has a chance to judge the credibility of that explanation, or when—as Zackowitz requires—all of this information is withheld?

1. The general rule and its foundations. Subject to certain exceptions to be explored below, the basic principle invoked by the majority in Zackowitz appears to enjoy universal acceptance: Other crimes (and indeed any other kind of evidence designed to show “bad character”) may not be introduced to show that the accused had an evil disposition and thus was more likely to have committed the offense charged.

Rules 403 and 404(b) of the Federal Rules of Evidence illustrate one rigorous statement of these principles:

What is the justification for these principles? McCormick states that such evidence “is not irrelevant, but in the setting of jury trial the danger of prejudice outweighs the probative value.” C. McCormick, Evidence §190, at 447 (2d ed. 1972). See also Michelson v. United States, 335 U.S. 469, 475-476 (1948): “The inquiry is not rejected because character is irrelevant; on the contrary it is said to weigh too much with the jury and to so overpersuade them as to prejudge one with a bad general record and deny him a fair opportunity to defend against a particular charge.” Undoubtedly, this statement holds true in a wide range of contexts. But there can be situations in which the danger of prejudice is arguably insufficient to justify the exclusion. Are there more basic reasons to exclude other-crimes evidence?

p. 32Consider how the criminal process would be affected if the prosecution were free to support its case by proving previous instances of a defendant’s criminal conduct. The accused, who is contesting the present indictment, may deny having committed the other crimes as well. And if the prosecution has not already obtained a conviction for the other crimes, the entire focus of the trial could be diverted by the dispute about whether those alleged crimes were in fact committed.

Even if the other crimes are easily proved by a record of prior convictions, the defendant may want to explain the background of the other offenses or claim extenuating circumstances. The vice here is not simply that time and attention may be diverted from the main issue in the case. It is also that a defendant should not be forever obliged to explain prior transgressions in order to dispel suspicions of further misconduct. A person who has suffered conviction and sentence is said to have “paid his debt to society”; the slate should be wiped clean. Cardozo alludes to these concerns when he states: “In a very real sense a defendant starts his life afresh when he stands before a jury, a prisoner at the bar. . . . Whether the explanation [for Zackowitz’s possession of the weapons] be true or false, he should not have been driven by the People to the necessity of offering it.”

Whether or not a defendant has already “paid his debt” for the prior offense, basic assumptions about criminal responsibility are tested when a trial focuses on the defendant’s general character rather than his behavior in a discrete situation. To be sure, a trial designed to determine a defendant’s responsibility for particular events often must explore many background circumstances, but ultimately the events themselves are at issue, the concrete behavior specified in the charges. The criminal trial usually is not viewed as a vehicle for passing judgment on the whole person. Again, Cardozo alludes to this principle: “Brought to answer a specific charge, [the defendant] had to defend himself against another, more general and sweeping[,] . . . pervasive and poisonous even if insidious and covert, that he was a man of murderous heart, or criminal disposition.” Consider Gerard E. Lynch, RICO: The Crime of Being a Criminal, Parts III & IV, 87 Colum. L. Rev. 920, 934-936 (1987):

Questions: To what extent are the values described fundamental to a just system of criminal law? Are they, for example, more important than the most p. 33accurate possible determination of the truth? Should they apply with as much force in proceedings to determine sentence (or the degree of the offense in Zackowitz) as they do when the issue is guilt versus innocence?48 Consider the extent to which these values in fact are respected, or flouted, by the doctrines of criminal law examined in the remainder of this section and throughout this book.

2. Exceptions to the rule. Despite the concerns mentioned by the court in Zackowitz, the rule against admitting other-crimes evidence is subject to a number of “exceptions.” For example, evidence that the defendant had previously stolen the pistol with which the victim was shot ordinarily will be admissible—not to show the defendant’s disposition to crime but to help identify him as the killer. Similarly, evidence that the defendant had previously committed a robbery witnessed by the victim ordinarily will be admissible—not to show propensity but to show his motive for this particular killing. See, e.g., White v. Commonwealth, 178 S.W.3d 470 (Ky. 2005).

Technically speaking, these situations do not involve exceptions to the rule of exclusion because, properly stated, that rule renders other-crimes evidence inadmissible only when offered “to prove a person’s character in order to show that on a particular occasion the person acted in accordance with [that] character.” Fed. R. Evid. 404(b). In other words, the rule itself does not bar the use of other-crimes evidence for some purpose other than that of suggesting that he acted in conformity with a bad character.

When evidence of other crimes is not offered to prove propensity, and therefore is not barred by Rule 404(b), the evidence nonetheless may run afoul of some other prohibition. In particular, evidence offered to prove identity or other nonpropensity matters will be barred by Rule 403 if its prejudicial effect substantially outweighs its probative value. Evidence, even when relevant, must be excluded when it “tends to subordinate reason to emotion in the factfinding process.” United States v. Queen, 132 F.3d 991, 997 (4th Cir. 1997). When other-crimes evidence is both highly relevant and highly inflammatory, courts face a dilemma; in such cases, the Supreme Court has held, judges must consider using factual stipulations or other alternative methods for conveying the essential facts to the jury in less prejudicial fashion. Old Chief v. United States, 519 U.S. 172 (1997).

It is not always easy to tell whether other-crimes evidence links the defendant to the crime in some legitimate way or whether it simply suggests a criminal propensity. The problem generates a large volume of litigation. For helpful discussion, see Paul S. Milich, The Degrading Character Rule in American Criminal Trials, 47 Ga. L. Rev. 775 (2013).

p. 34 3. Sex offenses. The Violent Crime Control and Law Enforcement Act of 1994 amended the Federal Rules of Evidence by adding the following provision (Fed. R. Evid. 413(a)):

An analogous provision (Fed. R. Evid. 414(a)) rendered the defendant’s commission of child molestation admissible in a prosecution for other acts of child molestation. Evidence of prior sex crimes can be admitted under Rules 413 and 414 only when the evidence also passes the Rule 403 requirement that its prejudicial effect not outweigh its probative value. United States v. Loughry, 660 F.3d 965 (7th Cir. 2011). With this safeguard in place, the admission of evidence of prior sex crimes does not amount to an unconstitutional denial of due process. People v. Falsetta, 986 P.2d 182 (Cal. 1999); Enjady v. United States, 134 F.3d 1427 (10th Cir. 1998). Nonetheless, some courts treat Rules 413 and 414 as “provid[ing] a presumption in favor of . . .  admission” for prior sexual offenses and rarely hold such evidence to be barred under Rule 403. E.g., United States v. Merz, 396 Fed. Appx. 838, 843 (3d Cir. 2010). Some have admitted evidence of prior sexual offenses even when those offenses were committed several decades before the incident in question. E.g., United States v. Benally, 500 F.3d 1085 (10th Cir. 2007). See Aviva Orenstein, Deviance, Due Process, and the False Promise of Federal Rule of Evidence 403, 90 Cornell L. Rev. 1487, 1519-1521 (2005).

What is the justification for treating prior sexual acts differently from other evidence of bad character or criminal propensity? Can an exception to the rule of exclusion be justified on the ground that sexual misconduct has greater predictive value than other misconduct? Or should the rule of exclusion be applied even more strictly, on the ground that evidence of prior sexual misconduct has especially strong prejudicial effects?

For in-depth examination of the issues, see Katharine K. Baker, Once a Rapist? Motivational Evidence and Relevancy in Rape Law, 110 Harv. L. Rev. 563 (1997). Professor Baker charges that “Rule 413’s proponents rely on antiquated notions of rapists as rare, depraved psychopaths who have some sort of perverse psychological need for sex.” The empirical evidence, she argues, indicates that convicted rapists are much less likely to repeat their crimes than are convicted larcenists and burglars. Professor Baker concludes that Rule 413 is misguided because it “relies on outmoded and demonstrably false stereotypes of who rapes, what rape is, and why rape might be different from other crimes.” Id. at 565, 578, 589.

In accord with Professor Baker’s analysis, the Judicial Conference of the United States, the policy-making body of the federal judiciary chaired by Chief Justice William Rehnquist, urged Congress to repeal the new rules. The Judicial Conference concluded that “the new rules, which are not supported by empirical evidence, could diminish significantly the protections that have safeguarded persons accused in criminal cases,” and noted that the rules posed a p. 35“danger of convicting a criminal defendant for past, as opposed to charged, behavior or for being a bad person.” The Judicial Conference also noted that its conclusions about the undesirability of Rules 413 and 414 reflected a “highly unusual unanimity” of the judges, lawyers, and academics who serve on its advisory committees. See 56 Crim. L. Rptr. 2139-2140 (Feb. 15, 1995). Nonetheless, Congress has declined to modify or repeal the rules.

Several states have adopted similar rules, modeled on Federal Rule 413, for admitting evidence of prior sex crimes, on the theory that a defendant’s “lustful disposition” is especially probative in sex-offense prosecutions. In most states, however, the usual rule excluding evidence of prior misconduct continues to apply in rape and child abuse trials, just as it does in other prosecutions, and in some states, legislative changes similar to Rules 413 and 414 have been held to violate the state constitution. E.g., State v. Gresham, 269 P.3d 207 (Wash. 2012); State v. Cox, 781 N.W.2d 757 (Iowa 2010). See David P. Bryden & Roger C. Park, “Other Crimes” Evidence in Sex Offense Cases, 78 Minn. L. Rev. 529 (1994); Sara Sun Beale, Prior Similar Acts in Prosecutions for Rape and Child Sex Abuse, 4 Crim. L.F. 307 (1993).

4. The impeachment exception. In all of the situations so far discussed, the question has been whether the prosecution can use evidence of other crimes as part of its case-in-chief. Even when the other-crimes evidence is clearly inadmissible for this purpose, if the accused chooses to testify in his defense, the prosecution generally will be permitted to ask about the other crimes when it cross-examines him. The prosecution will also be permitted to introduce other-crimes evidence in its rebuttal for purposes of impeaching the defendant’s testimony. Many states hold that such evidence is admissible automatically, regardless of whether its probative value outweighs its prejudicial effect. E.g., State v. Harrington, 800 N.W.2d 46 (Iowa 2011). In theory, the other-crimes evidence cannot be used to provide affirmative support for the prosecution’s case; it may be considered only for purposes of judging the credibility of the defendant’s testimony, and the jury will be so instructed.

The rationale of the impeachment exception appears to be that a person convicted of crime may be more likely to give false testimony than a citizen with a “clean” record. Whatever the soundness of this rationale when the previous misconduct involves perjury or similar crimes of dishonesty, its invocation in the case of other offenses can border on the absurd. In a prosecution for burglary, previous burglary convictions are not admissible to show the defendant’s disposition to commit this crime, but if the defendant claims to have been elsewhere at the time, the burglary convictions will generally be held admissible to show a possibility that the defendant may be disposed to perjury!

The impeachment exception is premised on the assumption that the jury will consider the other crimes only for the limited purpose of judging credibility and will not treat the other crimes as affirmative evidence of guilt. Is this a plausible assumption? Lawyers and social scientists have studied the question but have not reached definitive conclusions. The Note that follows p. 36explores the effectiveness of cautionary instructions and collects some of the available findings. The immediate problem is to understand how the other-crimes rules actually function, but questions about the effectiveness of jury instructions are central to understanding the actual impact of all the elaborate rules of criminal law and evidence that the jury is called upon to apply.

When other-crimes evidence has been introduced for impeachment purposes, a typical instruction to the jury might read as follows (1-7 Modern Federal Jury Instructions—Criminal p. 7.13 (2015)):

Such an instruction requires the jury to perform an intellectual task that is bound to run counter to the jury’s natural inclinations. Indeed the rule generally excluding other-crimes evidence is premised on the assumption that such evidence is relevant to guilt and is very difficult for the jury to keep in perspective once it becomes known. Thus jurors may be tempted to disregard the instruction, even if they are able to grasp the subtle distinction it asks them to draw. The problem arises over and over in administering the complex rules that ostensibly govern the criminal process. Yet just as the jury creates the need for many of these rules, the nature of the jury raises doubts about whether subtle or counterintuitive instructions actually affect the outcome of the case.

Experienced judges have expressed sharply divergent views about the effectiveness of jury instructions. In Spencer v. Texas, 385 U.S. 554, 565 (1967), the Supreme Court expressed its faith in “the ability of juries to approach their task responsibly and to sort out discrete issues given to them under proper instructions.” Others have been more skeptical. Justice Robert H. Jackson warned: “The naive assumption that prejudicial effects can be overcome by instructions to the jury . . . all practicing lawyers know to be unmitigated fiction.” Krulewitch v. United States, 336 U.S. 440, 453 (1949) (concurring opinion). In Dunn v. United States, 307 F.2d 883, 886 (5th Cir. 1962), the court was equally pessimistic:

p. 37The empirical evidence is mixed. A study of pretrial publicity found that volunteer “jurors” exposed to a sensational newspaper account of the case were much more likely to convict, but among jurors instructed to disregard the newspaper accounts, the difference in conviction rates disappeared.49 In contrast, researchers found that exposure to a legally inadmissible confession significantly increased the likelihood of a guilty verdict when other evidence was weak and that instructions to disregard the confession had no significant effect on the likelihood of conviction.50 Studies focusing on the impact of juror exposure to a defendant’s record of prior convictions often find that evidence of previous similar offenses substantially increases the likelihood of conviction and that cautionary instructions remove little or none of the prejudicial effect.51 Some studies suggest, ironically, that a cautionary instruction actually may increase the likelihood of conviction, especially when it is given at the end of the trial—possibly because the instruction reminds the jury about the negative information and makes it more salient just when deliberations are about to begin.52

To the extent that cautionary instructions fail to eradicate prejudicial effects, the result could be due, in part, to jurors’ inability to understand the subtle distinctions that such instructions sometimes require. Several studies have produced disturbing evidence that jurors often do not grasp the judge’s explanations of legal concepts. One study found that the average juror understands less than half of the judge’s instructions on the law.53

Whatever the available evidence may suggest about the usefulness of jury instructions, there clearly remain substantial doubts about whether instructions are as effective in practice as they are assumed to be in theory. What are the implications of this situation? Granted that jury instructions function imperfectly (at best), what is the alternative?

One pragmatic approach has been to work on improving the wording of instructions, so that their meaning is easier to understand.54 Beyond this, p. 38when a judge tells a jury to disregard evidence, should she explain why the law considers the evidence misleading? The standard cautionary instructions tend to be rather perfunctory. For ways to convey a more forceful message, see Albert W. Alschuler, Courtroom Misconduct by Prosecutors and Trial Judges, 50 Tex. L. Rev. 629, 652-654 (1972).

Can ameliorative reforms cure these difficulties, or is the basic problem more fundamental? Note that jury instructions are necessary in the first place only because we want citizens without special training to participate and at the same time, we want their decisions to conform to law. Do these partially inconsistent desires require a system that is simply too complex to function properly? This problem can be reconsidered in connection with the materials on other countries’ approaches to lay participation. See p. 62 infra.

1. The basis of the reasonable-doubt requirement. Is it clear, as Justice Harlan argues, that the “social disutility” of convicting an innocent person is always far worse than that of releasing a guilty person? William Blackstone went even further, asserting the often-repeated view that it is “better that ten guilty persons escape, than that one innocent suffer.” 4 William Blackstone, Commentaries on the Laws of England *352 (1765). Other judges and commentators have approved even higher ratios, such as one hundred to one or even more. See Alexander Volokh, n Guilty Men, 146 U. Pa. L. Rev. 173, 187-191 (1997). But how can we really be sure that it is less “costly” to release ten (or one hundred) suspected serial killers who are guilty than to convict one suspected serial killer who is innocent?

Although the reasonable-doubt requirement appears to enjoy broad support, some skeptics wonder where Blackstone and Justice Harlan can p. 40get the information necessary to calculate and compare these kinds of social costs. If the cost comparisons are arbitrary and indeterminate, does it follow that the reasonable-doubt requirement is not justified after all? Or does the reasonable-doubt requirement rest on a different principle? Regardless of any alleged balance of societal gains and losses, doesn’t every individual enjoy a fundamental right not to suffer punishment in the absence of reliable proof of fault?

2. How burden-of-proof problems arise. Problems relating to the reasonable-doubt standard normally arise first at the close of the prosecution’s case. If the judge decides that the evidence raises a reasonable doubt about guilt as a matter of law (an elusive concept explored in the next Note), the judge must direct a verdict for the defendant. The same problem of assessing the sufficiency of the evidence may be presented to the judge again at the close of all the evidence (when the defendant may again move for a directed verdict); it will be a central concern of the jurors in their deliberations; and it may arise again on appeal (when the defendant may seek reversal on the basis of insufficient evidence).

3. Reasonable doubt as a matter of law. Courts often have difficulty determining whether the evidence is insufficient as a matter of law. In a sense, the judges must give the defendant the benefit of the doubt. If there is a reasonable doubt, then a guilty verdict would seem improper. On the other hand, before taking a case away from the jury (or reversing a jury’s verdict), judges must resolve all evidentiary doubts against the proponent of the motion; in this sense the courts must give the prosecution the benefit of the doubt on the question whether its evidence does prove guilt beyond a reasonable doubt. The following comment from the opinion in Curley v. United States, 160 F.2d 229, 232-233 (D.C. Cir. 1947), helps clarify this elusive problem and explains the test that judges apply to determine evidentiary sufficiency:

4. Explaining reasonable doubt. The problem of evidentiary sufficiency arises not only for judges but for the jury, and the instructions accordingly must tell the jury how to evaluate the evidence. Empirical studies confirm that jurors convict more readily when instructed under a more-likely-than-not standard than when instructed under the reasonable-doubt standard,55 and courts must protect the jury against any instruction that might dilute the latter standard. Accordingly, a conviction must be reversed for error in explaining the reasonable-doubt standard to the jury, even when the appellate court does not find the evidence insufficient as a matter of law.

(a) In McCullough v. State, 657 P.2d 1157 (Nev. 1983), the trial judge had explained degrees of proof to the jury in terms of “a scale of zero to ten.” He placed the burden of persuasion in civil trials at just over five. He then described beyond a reasonable doubt as “seven and a half, if you had to put it on a scale.” The Nevada Supreme Court reversed, stating (id. at 1159): “The concept of reasonable doubt is inherently qualitative. Any attempt to quantify it may impermissibly lower the prosecution’s burden of proof, and is likely to confuse rather than clarify.” In In re As. H., 851 A.2d 456 (D.C. 2004), the only witness to identify the suspect as the perpetrator testified that her level of certainty was “seven or eight” on a scale of one to ten. Although acknowledging that it is “difficult (if not impossible) to place a meaningful numerical value on reasonable doubt,” the court found that “very few judges would consider an 80% probability sufficient” and that 70 percent was “altogether inadequate.” Accordingly, the court held (over one dissent) that as a matter of law “the evidence was insufficient to prove [guilt] beyond a reasonable doubt.” Id. at 457, 461.

(b) Courts often get into trouble when attempting to explain “reasonable doubt” in qualitative terms. In Cage v. Louisiana, 498 U.S. 39 (1990), the trial judge in a first-degree murder prosecution had instructed the jury that “a reasonable doubt [must be] founded upon a real tangible substantial basis and not upon mere caprice and conjecture. It must be such doubt as would give rise to a grave uncertainty. . . . A reasonable doubt is not a mere possible doubt. . . . What is required is not an absolute or mathematical certainty, but a moral certainty.” The Supreme Court reversed the conviction, holding that the references to “grave uncertainty” and “moral certainty” improperly diluted the Winship standard.

p. 42 (c) A traditionally accepted definition of reasonable doubt was the following, which was required by Cal. Penal Code §1096 (1994):

This language was drawn verbatim from an 1850 jury instruction given by Massachusetts Chief Justice Lemuel Shaw. See Commonwealth v. Webster, 59 Mass. 295, 320 (1850). But what is the meaning, to a modern juror, of “moral evidence” and “moral certainty”? Does the instruction meet the requirements of Winship and Cage v. Louisiana?

In Victor v. Nebraska, 511 U.S. 1 (1994) (decided with a companion case from California), the Court distinguished Cage and upheld the constitutionality of the California instruction. The Court conceded that “the phrase ‘moral evidence’ is not a mainstay of the modern lexicon,” and that “moral certainty is ambiguous” but concluded that the instruction as a whole gave sufficient content to the reasonable-doubt requirement. Several concurring justices agreed that use of the nineteenth-century phrases was not unconstitutional but urged states to choose more modern language. California responded the following year by deleting the phrase “moral certainty”; the current statute describes reasonable doubt as a situation in which jurors “cannot say they feel an abiding conviction of the truth of the charge.” Cal. Penal Code §1096 (2016). In some states, the “moral certainty” language is still used. E.g., 7 Tenn. Prac. Pattern Jury Instr. T.P.I.-Crim. 2.03 (2016), explaining that “[a]bsolute certainty of guilt is not demanded . . . , but moral certainty is required, and this certainty is required as to every element of proof necessary to constitute the offense.” Massachusetts, where the “moral certainty” language originated, continues to use it but now explains to the jury that moral certainty means “the highest degree of certainty possible in matters relating to human affairs.” Commonwealth v. Russell, 23 N.E. 3d 867, 878 (Mass. 2015).

Recent studies indicate that among various formulations, jurors told that they must be “firmly convinced” were most likely to differentiate between strong and weak cases, acquitting in the latter. But even when given the “firmly convinced” language, “more than one-third of the [mock] juries voted to convict when the prosecution’s evidence did not even satisfy a preponderance of the evidence standard.” See Stoltie v. California, 501 F. Supp. 2d 1252, 1261 (C.D. Cal. 2007). Many experts believe “[c]olloquial words that describe subjective certainty will likely be more effective in explaining to the jury the truly high standard of proof.” Id. at 1262. But to date, courts have not required more emphatic explanations of this sort.

(d) Because reasonable doubt is difficult to explain correctly, several courts consider it preferable to give the jury no explanation at all. E.g., United States v. Hornsby, 666 F.3d 296, 310-311 (4th Cir. 2012). In United States v. p. 43Walton, 207 F.3d 694 (4th Cir. 2000), a jury that had received no explanation of reasonable doubt sent the judge a note asking him to define it. After the judge refused, the jury convicted. The Fourth Circuit affirmed, stating that trying to explain things will “confuse rather than clarify.” Is this an appropriate solution to the problem?

1. Duncan in context. Justice White mentions, early in his opinion, that Duncan’s cousins had recently transferred to a previously all-white high school where “racial incidents” had been reported. That description of the circumstances was a colossal, and no-doubt deliberate, understatement. The previous year, the U.S. attorney general had declared Plaquemines Parish one of the nine most discriminatory counties in the nation, and in 1966, at the time of the incident, the parish was in the midst of a ferocious school desegregation battle, pitting Justice Department attorneys, civil rights lawyers, and the federal courts against Leander H. Perez, Sr., a political figure and ardent segregationist (known locally as “the Bonaparte of the Bayou”) who had ruled the parish with an iron fist for decades. Perez had hand-picked all judges and other officials in the parish, and he tolerated no opposition. Many civil p. 57rights activists who dared to enter the parish were arrested as “outside agitators.” When the civil rights lawyer who had taken Duncan’s case appeared at the courthouse to file a motion needed for the appeal to the U.S. Supreme Court, the Plaquemines sheriff arrested him on a charge of practicing law without a license and confiscated the briefcase that contained all of Duncan’s appeal papers. For a fascinating, in-depth description of this background, see Nancy J. King, Duncan v. Louisiana: How Bigotry in the Bayou Led to Federal Regulation of State Juries, in Criminal Procedure Stories (C. Steiker ed., 2006).

Why didn’t the Court make more of these local circumstances? The political situation in Plaquemines Parish seems to afford a perfect opportunity to make vivid the value of the jury as a check on official power. Would an emphasis on these facts have made the Court’s opinion more persuasive? Or would such an emphasis detract from the Court’s effort to establish a more general principle—that the jury should be considered an important safeguard against abuse even in a well-functioning democracy?

2. The scope of the right to jury trial. Justice Felix Frankfurter once wrote that “[n]o changes or chances can alter the content of the verbal symbol of ‘jury’—a body of twelve men who must reach a unanimous conclusion if the verdict is to go against the defendant.” Rochin v. California, 342 U.S. 165, 170 (1952). Except for the limitation to males, which had passed away long before Justice Frankfurter reiterated it, this statement expressed a nearly universal view about what was meant by a jury. Nevertheless, in Williams v. Florida, 399 U.S. 78, 86 (1970), the Court said that the decision to fix the size of the jury at 12 “appears to have been a historical accident, unrelated to the great purposes that gave rise to the jury” and held that a 6-member jury satisfied the constitutional requirement.57 In Apodaca v. Oregon, 406 U.S. 404 (1972), the Court held that unanimity was not required in state criminal trials, so long as a substantial majority of the jury supports the verdict. The Court in that case upheld guilty verdicts obtained by 11–1 and 10–2 votes, without ruling explicitly on whether a smaller majority could also be sufficiently substantial.58

Duncan did not resolve the question of what may be deemed a “petty offense,” for which the Sixth Amendment jury trial guarantee would be inapplicable. In Baldwin v. New York, 399 U.S. 66 (1970), the Court held that no offense may be deemed petty where imprisonment for more than six months is authorized. In such cases, a defendant has a constitutional right to jury trial, whether or not imprisonment is in fact likely to be imposed.

p. 58 3. The representative jury. Contrary to popular folklore, there is no requirement that a defendant be tried by a jury of “his peers.” Nor is there any requirement that the trial jury of 6 or 12 reflect the demographic character of the locality. But the Supreme Court has held that under the Sixth Amendment, the “venire”—the panel of potential jurors from which the trial jury is drawn—must reflect “a fair cross section of the community.” Taylor v. Louisiana, 419 U.S. 522, 537 (1975). The Court has treated the “fair cross section” requirement as a means to ensure the impartiality of the jury, but has not recognized an independent Sixth Amendment right to a jury that represents the community in any broader sense. See Holland v. Illinois, 493 U.S. 474 (1990); Nina W. Chernoff, Wrong About the Right: How Courts Undermine the Fair Cross-Section Guarantee by Confusing it with Equal Protection, 64 Hastings L.J. 141 (2012); Laura G. Dooley, The Dilution Effect: Federalization, Fair Cross-Sections, and the Concept of Community, 54 DePaul L. Rev. 79 (2004).

Once the venire is assembled, potential jurors who know the defendant, the victim, or a witness can be challenged “for cause” and removed from the pool. The opposing attorneys are then allowed “peremptory challenges,” which permit them to remove a certain number of potential jurors without giving any reason, simply because they suspect that the potential juror may be unsympathetic to their side. In Batson v. Kentucky, 476 U.S. 79 (1986), and J.E.B. v. Alabama, 511 U.S. 127 (1994), the Supreme Court held that attorneys cannot use peremptory challenges to deliberately exclude potential jurors on grounds of race or gender. But the requirement is difficult to enforce because a peremptory challenge can pass muster when the attorney can offer a plausible race-neutral and gender-neutral explanation for making it. As a result, the trial jury ultimately empanelled often will differ markedly from the character of the community as a whole, even when the venire was well balanced. An experienced federal judge reported that in Boston, “the vast majority of juries in federal court . . . did not have a single African-American member.” Nancy S. Gertner, 12 Angry Men (and Women) in Federal Court, 82 Chi.-Kent L. Rev. 613, 619 (2007).

4. The effect of jury trial on the criminal justice system. (a) The entire texture of the trial is influenced by the existence of the jury. Instead of addressing arguments to one law-trained person, the lawyers address themselves to 12 lay people. Lawyers are often convinced that nonlegal factors will influence the jury even when those factors are not technically “relevant,” and as a result they often attempt, with some success, to find indirect ways to make the jury aware of such matters.

(b) Because lay people may not assess items of proof as carefully as trained lawyers, Anglo-American law includes elaborate rules for excluding certain evidence at the trial.

(c) The judge is judge of the law; the jury decides questions of fact. The legal system must characterize the nature of a given question: Is it a question of law or of fact?

p. 59 (d) Judges must formulate for jurors’ use an acceptable statement of applicable legal rules, though these rules are often extremely complex. When significant errors are made in stating these rules, a conviction must be reversed, and a new trial must be held.

5. The policies served (and disserved) by jury trial. The Court in Duncan summarizes the principal reasons why it regards the availability of trial by jury as an essential component of fair procedure. Is it clear that the advantages of jury trial outweigh its costs or that experimentation with different kinds of factfinding procedures should be considered intolerable? Consider the following comments:

1. Effects in practice. Empirical studies have probed the impact of nullification instructions. A fundamental question underlying the debate is whether jurors who nullify are reflecting the community sense of justice, p. 67or instead (as the Dougherty court feared) are responding to vengeful instincts and emotional biases triggered by characteristics of the victim or the defendant. A further question, then, is whether explicit instructions about the nullification power will exacerbate these possibilities. One empirical study probing these issues65 considered the effect of instructing mock juries that “nothing would bar them from acquitting the defendant if they feel that the law . . . would produce an inequitable or unjust result.” In a case involving a nurse tried for the “mercy” killing of a terminally ill cancer patient, mock juries given the nullification instruction were, predictably, less likely to convict than mock juries not given the instruction. But, unexpectedly, the instructed mock juries were more likely to convict in a homicide case involving a drunk driver who struck and killed a pedestrian. In deliberations, mock juries spent less time discussing the evidence and more time discussing the defendant’s character when a nullification instruction had been given. Another study found that in cases where the fairness of the underlying law was a concern, jurors given negative information about the victim were less likely to convict when given a nullification instruction, whereas negative information about the victim did not affect the likelihood of conviction when jurors were given standard jury instructions, without reference to nullification.66

2. Current state of the law. The federal courts and nearly all the states follow Dougherty and refuse to permit instructions informing the jury of its nullification power. See, e.g., United States v. Moran-Toala, 726 F.3d 334, 343 (2d Cir. 2013); State v. Paul, 104 A.3d 1058 (N.H. 2014). In State v. Kelly, 727 S.E.2d 912, 915 (N.C. Ct. App. 2012), the court declared that granting a right to nullification instructions “would lead to chaos.” In contrast, in United States v. Folse, 2015 U.S. Dist. LEXIS 176610, at *54 (D.N.M. Nov. 9, 2015), the court acknowledged that precedent precluded a nullification instruction but nonetheless, after review of the early history, concluded that “the current prohibitions against any mention of the jury’s ability to mitigate or nullify its verdict are inconsistent with the Framers’ intent.” For further development of that argument, see Clay S. Conrad, Jury Nullification 13-115 (2014).

In three states, the approach rejected in Dougherty survives through constitutional provisions that the jury shall be the judge of the law as well as the fact. Ga. Const. art. 1, §1, §11(a); Ind. Const. art. 1, §19; Md. Const., Decl. of Rights, art. 23. In Georgia, however, courts have confined the provision—for example, by upholding a charge that the jurors are the judges of the law but are obliged to apply the court’s instructions to the facts and by forbidding defense counsel to argue to the jury that it should disregard the law. See Newsome v. State, 747 S.E.2d 99 (Ga. Ct. App. 2013). Indiana and p. 68Maryland practice is summarized in Kenneth Duvall, The Contradictory Stance on Jury Nullification, 88 N.D. L. Rev. 409, 414 (2012).

3. Is nullification ever desirable? The Dougherty court appears to believe that the jury’s nullification power is desirable, so long as it is not exercised too often. Compare State v. Ragland, 105 N.J. 189, 519 A.2d 1361, 1371-1372 (1986):

4. Other mechanisms of “arbitrary” leniency. The Ragland court rests part of its case against jury nullification on the ground that this power “arbitrarily, to pick and choose” defendants for unexplained leniency exists “[a]bsolutely nowhere [else] in the system.” But at the outset of this book we noted the importance of discretion throughout the criminal justice system. In particular, the judgments of police and prosecutors about whether and what to charge are largely unguided and uncontrolled. See p. 8, supra, and the flow chart on p. 19, supra. Another important source of discretionary leniency is the power of state governors and the President to grant executive clemency, a power that is likewise unrestricted and therefore in essence “arbitrary.” In Ex parte Garland, 71 U.S. (4 Wall.) 333, 380 (1867), the Supreme Court declared that “[t]he benign prerogative of mercy reposed in [the President] cannot be fettered by any legislative restrictions.” Given the p. 69phenomenon of pervasive discretion (which we examine in detail in Chapter 10, infra), does jury nullification really deserve special skepticism? Is unexplained leniency granted by a citizen group serving on a one-time basis more trustworthy or more problematic than unexplained leniency granted by elected officials accountable to voters and the political process?

Professor Andrew D. Leipold considers nullification undesirable, and he challenges the Ragland court’s premise that the nullification power is unavoidable. He would restrict the de facto nullification power by authorizing prosecution appeals (a view that would require reversal of current double jeopardy case law);67 at the same time, he would permit explicit instructions authorizing nullification in cases involving de minimis harms, and would bar prosecution appeal from an acquittal on that ground (if made explicit by a jury’s special verdict). See Leipold, Rethinking Jury Nullification, 82 Va. L. Rev. 253 (1996).

A different view is developed in Sherman J. Clark, The Courage of Our Convictions, 97 Mich. L. Rev. 2381 (1999). Professor Clark argues that jury trials “serve as a means through which we as a community take responsibility for—own up to—inherently problematic judgments regarding the blameworthiness or culpability of our fellow citizens.” Id. He suggests, accordingly, that “procedures governing the criminal jury trial [should] engender in jurors a sense of personal responsibility for the fate of the accused.” Id. at 2382. He recommends that jurors be instructed in a way that stresses their responsibility for their decision and makes them aware of their nullification power, without expressly encouraging them to use it—for example, by telling the jurors (id. at 2446) that “the responsibility for this decision is entirely yours and . . . you will not be required to explain or justify your verdict except to your own conscience.”

5. The legality principle. A traditional principle of criminal law—also a constitutional mandate under the requirement of “due process of law”—is that punishment can be imposed only for violation of clear standards promulgated in advance of the defendant’s conduct. See pp. 158-202 infra. Suppose that the jurors are instructed, over the defendant’s objection, that they are the judges of the law as well as the facts. Does such an instruction subject the defendant to capricious judgment and violate his right to be tried in accordance with ascertainable law, with guilt established beyond a reasonable doubt? Or are due process concerns met by the trial judge’s power to set aside a conviction not supported by sufficient evidence? See State v. Waine, 122 A.3d 294, 301 (Md. 2015).

6. Biased nullification. Consider Tania Tetlow, Discriminatory Acquittal, 18 Wm. & Mary Bill Rts. J. 75, 75 (2009):

Question: If Professor Tetlow is right, what can be done to preserve nullification as a safeguard against government overreach while countering its dangers for underprotected victims?

7. Constraining nullification. If jury nullification is usually or (as the Ragland court argues) always undesirable, how far can courts go to discourage it? In a civil suit, the judge can set aside a jury verdict for one party and enter judgment notwithstanding the verdict (often called “judgment n.o.v.”) for the other side. But in a criminal case, a judgment n.o.v. or directed verdict for the prosecution violates the defendant’s Sixth Amendment right to trial by jury. Connecticut v. Johnson, 460 U.S. 73, 84 (1983); United States v. Salazar, 751 F.3d 326, 334 (5th Cir. 2014). Are other judicial efforts to prevent nullification similarly barred by the Sixth Amendment? Consider these situations:

(a) In People v. Fernandez, 31 Cal. Rptr. 2d 677 (Ct. App. 1994), the jurors interrupted their deliberations to send the judge a note in which they stated their belief that the defendant had committed battery with serious bodily injury and then asked whether they nonetheless had the power to acquit on that charge and instead return a verdict of guilty on the lesser offense of assault. The trial judge simply replied, “No.” The court of appeal held that this response was not error.

(b) In State v. Smith-Parker, 340 P.3d 485, 499 (Kan. 2014), the defendant was convicted of murder after the judge instructed the jury: “If you do not have a reasonable doubt from all the evidence that the State has proven murder in the first degree on either or both theories, then you will enter a verdict of guilty” (emphasis added). The Kansas Supreme Court reversed, holding that the “wording at issue here—‘will’—fl[ies] too close to the sun of directing a verdict for the State. A judge cannot compel a jury to convict, even if it finds all elements proved beyond a reasonable doubt.” Id. at 507. Compare United States v. Folse, 2015 U.S. Dist. LEXIS 176610, at *7 (D.N.M. Nov. 9, 2015), where the jury convicted defendant of murder after being instructed: “You have no right . . . to question the wisdom or correctness of any rule I may state to you. You must not substitute or follow your own p. 71notion or opinion as to what the law is or ought to be. It is your duty to apply the law as I explain it to you, regardless of the consequences.” Id. at *7. Is the instruction consistent with Dougherty? Should the conviction in Folse be upheld on appeal?

(c) United States v. Thomas, 116 F.3d 606 (2d Cir. 1997), was a prosecution of several African-American defendants charged with conspiracy to distribute cocaine. During deliberations, several jurors informed the judge that Juror No. 5, the sole African American on the panel, was refusing to follow the judge’s instructions, was calling his fellow jurors “racists,” and was adamantly holding out for acquittal. The judge interrupted the deliberations and interviewed each of the jurors. According to several of them, Juror No. 5 had said he thought the government’s evidence was unreliable, and Juror No. 5 told the judge that he needed “substantial evidence” of guilt. But several other jurors testified that Juror No. 5 had said he favored acquittal because the defendants were black and had committed the alleged crimes out of economic necessity. The trial judge accepted the latter account and dismissed Juror No. 5 on the ground that he would not convict “no matter what the evidence was” and had “preconceived . . . economic [or] social reasons [for acquittal] that are totally improper.” The remaining jurors then unanimously convicted.

On appeal, the court, per Judge José Cabranes, “categorically reject[ed] the idea that jury nullification is desirable or that courts may permit it to occur when it is within their authority to prevent it.” The court held that the Constitution permits removal whenever there is unambiguous evidence of a juror’s refusal to follow the judge’s instructions. The court reversed the convictions, however, because Juror No. 5’s intentions were ambiguous and there was “some possibility” he had based his vote on his view of the evidence.

Roughly in accord with the Thomas standard, most of the recent decisions hold that it is error to remove an allegedly recalcitrant juror if there is “any reasonable possibility” that the juror is following the judge’s instructions. State v. Morfin, 287 P.3d 600, 605 (Wash. Ct. App. 2012). In United States v. Abbell, 271 F.3d 1286, 1302 (11th Cir. 2001), the court described this as “basically a ‘beyond reasonable doubt’ standard.” Nonetheless, courts often find that this tough requirement has been met and accordingly uphold decisions to remove a juror who has been holding out for acquittal. E.g., United States v. Godwin, 765 F.3d 1306 (11th Cir. 2014). And in Johnson v. Williams, 133 S. Ct. 1088, 1098 (2013), the Court upheld a more permissive California approach that allows a juror to be discharged when inquiry reveals a “demonstrable reality” that the juror is unwilling or unable to deliberate.

Questions: If jury departures from the law are so unequivocally bad, why should the defendant have a constitutional right to jury trial in the first place? Was Duncan wrongly decided? Conversely, if the jury’s equity-dispensing function is central to its constitutional role, how can it be proper for a trial judge to remove jurors who reveal their intent to exercise that function? And what effect will the existence of such a removal power have on jurors’ ability p. 72to express their views candidly during deliberation? For discussion of the issues, compare Nancy S. Marder, The Myth of the Nullifying Jury, 93 Nw. U. L. Rev. 877, 947-952 (1999) (criticizing Thomas), with Nancy J. King, Silencing Nullification Advocacy Inside and Outside the Courtroom, 65 U. Chi. L. Rev. 433, 438-491 (1998) (supporting the Thomas approach).

(d) The Fully Informed Jury Association (FIJA) is a Montana-based organization supported by a politically diverse array of protest groups, including anti-abortion activists, supporters of alternative medicines, and opponents of laws regulating firearms, marijuana, motorcycle helmets, prostitution, and the right to die. See King, supra at 434; Marder, supra at 942. Through a national newsletter, website, and handbills distributed at courthouses, FIJA seeks to spread awareness of the nullification power and encourage its use. At many courthouses throughout the country, FIJA distributes pro-nullification literature to potential and actual jurors, and encourages them to call an 800 number to hear a recorded message describing what one juror discovered were “more rights than what was read to me by the judge.”68

Recall that in Dougherty, the majority considered formal nullification instructions unnecessary, in part because “the jury gets its understanding . . .  from more than one voice”; the court concluded that awareness of the nullification power can best come from “informal input.” Are FIJA’s activities therefore legitimate and perhaps even desirable? Many judges and other court officials apparently do not think so. FIJA activists who contact jurors or engage in leafleting at the courthouse have been charged with jury tampering, obstruction of justice, and contempt of court, even though FIJA’s informational materials are not alleged to be incorrect or misleading. An Alaska case involved an FIJA activist who approached three potential jurors and urged them to call a toll-free number to hear a recorded message about the jury’s powers. He was convicted of jury tampering under a state statute that prohibits “directly or indirectly communicat[ing] with a juror . . . with intent to . . . influence the juror’s vote, opinion or decision.” The U.S. Court of Appeals rejected his First Amendment challenge to the conviction. Turney v. Pugh, 400 F.3d 1197, 1199 (9th Cir. 2005). More recently, federal authorities filed jury tampering charges against Julian Heicklen, a 79-year-old retired university professor who allegedly stood on the plaza outside the U.S. Courthouse in Manhattan handing out brochures to passers-by (some of whom could have been potential jurors), accurately explaining the jury’s power to acquit in cases where they disagree with the law. The government used an undercover agent to secretly record a conversation in which Heicklen allegedly said “I’m not telling you to find anybody not guilty. But if there is a law you think is wrong then you should do that.” Prosecutors insisted that his speech was not protected by the First Amendment. Lawyers helping to defend him commented that the prosecution would raise awareness of the nullification power and “surely convert more to the cause than poor Mr. Heicklen ever p. 73could on his own.” See Benjamin Weiser, Prosecution Explains Jury Tampering Charge, N.Y. Times, Nov. 27, 2011. Ultimately, the trial court dismissed the jury-tampering charge, holding that the statute banned communication on an “issue or matter” before the jury, and that brochures merely describing a juror’s duties did not involve an “issue or matter” unless directed to a specific case. United States v. Heicklen, 858 F. Supp. 2d 256, 275 (S.D.N.Y. 2012).

Municipalities have responded to FIJA’s efforts by enacting laws to more tightly restrict contact with potential jurors in or near the courthouse.69 Freedom of speech principles do not prohibit reasonable restrictions on advocacy within a fixed distance from decision-making sites like courthouses or voting booths.70 But First Amendment considerations aside, why should it be a crime to give a juror truthful information about her rights and responsibilities? If society wants to block both formal and informal sources of information about the jury’s equity-dispensing role, why bother having a jury at all? For analysis of judicial efforts to silence nullification advocacy, see King, supra Note 7(c), at 492-499.

Professor Andrew D. Leipold raises an additional concern. If jurors are supposed to decide whether application of the law is unfair in a given case, they “would probably want to know [a host of details, including] whether the defendant is contrite[,] . . . whether he had a criminal record, and if so, how serious were his prior crimes. [They] might wonder how the prosecution enforces this crime against others: are African-Americans disproportionately targeted[?] . . . The problem is that almost none of this information is admissible at trial.” Leipold, Race-Based Jury Nullification: Rebuttal (Part A), 30 J. Marshall L. Rev. 923, 923-924 (1997). Can we assume that the jury will glean this information from informal sources? And if nullification is a legitimate option, does such information become legally relevant, so that it should be admissible at trial? What are the perils of that approach?

8. Can race-based nullification be benign? In Detroit, Washington, D.C., parts of New York City, and several other urban centers, observers have claimed that jury nullification is becoming more common, especially in drug prosecutions involving African-American defendants. See Marder, supra Note 7(c), at 899-901. Consider the following comments:

Because Duncan emphasizes that a central jury function is to afford protection against the “compliant, biased, or eccentric judge,” it might seem that the constitutional right to jury trial should include a right to have the jury decide not only matters of guilt and innocence but the offender’s punishment. In capital cases, nearly all states do give the jury this broader responsibility; we explore the jury’s special role in this context in Chapter 5, pp. 567-602 infra. But for other sentencing matters, only six states allow for punishment to be determined largely by the jury. In all the others, the sentencing decision is left to the judge, and this arrangement is normally not thought to violate Sixth Amendment requirements. Why not? Why should decisions about the sentence be treated differently from decisions about whether to convict? And what freedom should states have to shift factual questions from the guilt phase, where the jury must resolve them, to the sentencing phase, where the prosecution needs only to convince a single judge? Consider the Notes that follow.

1. Jury sentencing. In Arkansas, Kentucky, Missouri, Oklahoma, Texas, and Virginia, defendants who plead guilty or opt for a bench trial are sentenced by a judge, but those who elect to go to trial before a jury are sentenced by the same jury, usually in a “bifurcated” proceeding—the trial on guilt or innocence occurs first, and if the jury convicts, it then hears further evidence and retires to deliberate separately on the question of punishment. (The principal exception is Oklahoma, where first offenders are tried and sentenced in a unitary proceeding.)

At the penalty phase, sentencing juries typically are not given guidelines of any sort. Instead, they are simply asked to pick any sentence within the authorized statutory range. (In contrast, in many jurisdictions guidelines and presentence reports shape the sentencing decisions of judges, as discussed in Chapter 10, infra.) As a result, jury sentencing produces enormous variation in the sentences imposed on similarly situated offenders. See Nancy J. King & Rosevelt L. Noble, Felony Jury Sentencing in Practice: A Three-State Study, 57 Vand. L. Rev. 885 (2004); Jenia Iontcheva, Jury Sentencing as Democratic Practice, 89 Va. L. Rev. 311 (2003). In addition, sentencing juries typically have fewer options for leniency than are available when the sentence p. 77is fixed by the judge. For example, in several of the jury-sentencing states, judges can grant probation or a suspended sentence, but juries cannot. For this reason and others, studies suggest, defendants sentenced by a jury typically receive more severe sentences than similarly situated defendants sentenced by a judge. See Nancy J. King & Rosevelt L. Noble, Jury Sentencing in Non-Capital Cases: Comparing Severity and Variance with Judicial Sentences in Two States, 2 J. Empirical Legal Stud. 331 (2005).71

Despite these features (or perhaps because of them), state officials have shown little enthusiasm for reforms that would permit more leniency or consistency in jury sentencing, through sentencing guidelines and other devices now widely used to promote uniformity in sentencing by judges. Prosecutors and legislators in jury-sentencing states seem to believe that the greater severity and unpredictability of jury sentencing is an important factor in inducing guilty pleas, so that cases can be disposed of quickly and cheaply. Professor Nancy King concludes that “the wild card aspect of jury sentencing helps to funnel defendants to guilty pleas and bench trials. . . . For criminal justice insiders, the unpredictability of jury sentencing is a blessing, not a curse; the more freakish, the better.” Nancy J. King, How Different Is Death? Jury Sentencing in Capital and Non-Capital Cases Compared, 2 Ohio St. J. Crim. L. 195, 198 (2004). Yet courts have consistently rejected the argument that unguided jury discretion in non-capital cases violates due process. E.g., Torres v. United States, 140 F.3d 392, 397 (2d Cir. 1998).

Do these problems suggest that at the sentencing stage, the jury cannot serve as an effective equity-dispensing mechanism for the benefit of defendants? Or do they simply show that jury sentencing has not been permitted to fulfill its potential?

2. Judicial sentencing: What is the jury permitted to know? Does it follow from Dougherty that a judge not only may refuse to inform the jury of its nullification power but also may refuse to inform the jury about the severity of the sentence that a defendant faces upon conviction? The question has become increasingly important with the proliferation of mandatory minimum sentencing laws, which often dictate long terms of imprisonment for things like possession of drugs, even when the defendant is a first offender. Yet nearly all courts hold that because the jury’s role is solely to determine the facts relevant to guilt, the jury has no legitimate concern with the consequences of a conviction. E.g., Shannon v. United States, 512 U.S. 573 (1994). For criticism of this approach, see Jeffrey Bellin, Is Punishment Relevant After All? A Prescription for Informing Juries of the Consequence of Conviction, B.U. L. Rev. 2223 (2010); Michael T. Cahill, Punishment Decisions at Conviction: Recognizing the Jury as Fault-Finder, 2005 U. Chi. Legal F. 91.

p. 78United States v. Polizzi, 549 F. Supp. 2d 308 (E.D.N.Y. 2008), illustrates another challenge to the prevailing view. The defendant was convicted on 12 counts of receiving child pornography. The jury rejected his insanity plea despite evidence that he had suffered vicious sexual abuse as a child, had nonetheless established a home with a loving family, and had no prior criminal history. After the verdict, a poll of the jurors revealed that, had they been aware of the applicable five-year mandatory minimum sentence, they would not have convicted, because they thought the defendant should receive treatment and close supervision to prevent a recurrence, not a long prison term. Noting that a properly informed jury would likely have deadlocked or found the defendant not guilty by reason of insanity, the district judge awarded a new trial. Based on an extensive review of the history of jury trial, the judge held that under these circumstances, the failure to inform the jury of the mandatory minimum sentence prior to its deliberations deprived the defendant of his constitutional right to jury trial, as that right had been understood at the time of the framing of the Bill of Rights. However, the Court of Appeals reversed, holding that absent unusual circumstances, it is an abuse of discretion for a trial judge to inform the jury of the sentencing consequences of its verdict. United States v. Polouizzi [a/k/a/ Polizzi], 564 F.3d 142 (2d Cir. 2009).72 Compare United States v. Datcher, 830 F. Supp. 411, 414-418 (M.D. Tenn. 1993), where another district judge, protesting against the prevailing bar on instructions about sentencing consequences, commented:

3. Judicial sentencing: When can factual questions be shifted from the jury’s domain to that of the judge? Duncan protects the defendant’s right to have the jury determine, under a strict beyond-a-reasonable-doubt standard, all facts necessary to constitute the crime charged—that is, all “elements” of the offense. But a single judge, acting under a preponderance-of-the-p. 79evidence standard, can determine all other facts relevant to the level of punishment—that is, all “sentencing factors.” Years of imprisonment can therefore turn on whether the fact is characterized as an “element” of the offense or merely as a “sentencing factor.” For discussion of the case law on this issue, see p. 51 Note 3, supra. Consider to what extent the “sentencing factor” option (together with the rule against informing the jury about the sentencing consequences of a conviction (Note 2 supra)) enables the legislature to defeat the jury’s constitutionally protected functions identified in Duncan—the functions of restraining overly harsh laws and preventing prosecutors from using generally reasonable laws in an oppressive manner. For discussion, see Rachel E. Barkow, Recharging the Jury: The Criminal Jury’s Constitutional Role in an Era of Mandatory Sentencing, 152 U. Pa. L. Rev. 33 (2003).

1. The problem and the prevailing solution. When a prosecution involves several separate counts, it sometimes happens that the jury’s verdict on one count will be irreconcilably in conflict with its verdict on another count. Consider, for example, DeSacia v. State, 469 P.2d 369 (Alaska 1970). The defendant’s reckless driving forced another vehicle off the road, and both the driver (Hogan) and passenger (Evangelista) in the other vehicle were killed. The defendant was prosecuted on two counts of manslaughter. The jury convicted on the count charging manslaughter of Evangelista but acquitted on the count charging manslaughter of Hogan. Because the defendant’s conduct had endangered the two victims in precisely the same way, he was in principle guilty of manslaughter either in both cases or in neither. Should the inconsistent verdicts nevertheless be allowed to stand?

In the DeSacia case, the court noted that the verdict of acquittal was final; any relitigation of that count would violate the Double Jeopardy Clause. But the court set aside the conviction on the other count, to assure that the conviction was not the product of jury confusion or irrationality. The court therefore remanded for a new trial on the count relating to the death of Evangelista. Similarly, in Price v. State, 405 Md. 10, 23-29 (2008), the court acknowledged that inconsistent verdicts had been accepted in the interests of lenity or compromise, but held that such verdicts, because they are contrary to the law, would no longer be tolerated.

Although several American jurisdictions refuse to accept inconsistent verdicts, see McNeal v. State, 44 A.3d 982 (Md. 2012), the federal courts and the great majority of the states permit the inconsistent conviction to stand. See Priest v. State, 879 A.2d 575 (Del. 2005); People v. Jones, 207 Ill. 2d 122, 133 (2003); Eric L. Muller, The Hobgoblin of Little Minds? Our Foolish Law of Inconsistent Verdicts, 111 Harv. L. Rev. 771, 787-788 (1998). Which approach seems more faithful to the premises of jury trial? If the jury is expected to give voice to the rough common sense of the community, isn’t the DeSacia court’s desire for consistency out of place? Do inconsistent p. 80verdicts facilitate the exercise of the jury’s leniency or do they encourage compromise convictions on counts about which the jury may not really be persuaded beyond a reasonable doubt? In United States v. Powell, 469 U.S. 57, 65 (1984), the Supreme Court noted:

2. Criticism of the prevailing acceptance of inconsistent verdicts. The Powell Court’s assumptions have drawn sharp criticism. Professor Leipold notes that “[f]or every case where the jury extends mercy to a deserving defendant, there may well be another (or two, or five others) where the verdict is based on improper considerations.” Andrew Leipold, Rethinking Jury Nullification, 82 Va. L. Rev. 253, 304 (1996). Professor Muller, supra Note 1, at 795, argues that “by refusing to disturb inconsistent verdicts, the Court is buying a chance for [leniency] at the price of protecting jury confusion, mistake, and compromise.” He concludes that courts should set aside inconsistent convictions unless the evidence supporting the conviction is so overwhelming that any jury error can be deemed harmless.

Questions: What if the prosecution seeks an instruction, in a case like DeSacia, that the jury must either acquit on both counts or convict on both counts? Does the result in DeSacia in effect require such an instruction in future cases? If so, would the defendant then be justified in claiming that the instruction improperly constrained the jury’s equity-dispensing function?

In connection with these problems, consider whether it was entirely a coincidence that the DeSacia jury convicted for the death of Evangelista, who was a passenger, but acquitted for the death of Hogan, the other driver. As we shall see, contributory negligence is not a defense in a criminal prosecution, see p. 495 infra, but the Kalven and Zeisel study, p. 62 supra, at 242-257, showed that jury nullification often occurred when contributory fault by the victim was involved. If this factor played a role in the acquittal with respect to Hogan, then wasn’t the jury—to quote from Duncan v. Louisiana, p. 55 supra—“serving some of the very purposes for which [it was] created”? Or are the dangers identified by Professor Muller (jury confusion, mistake, inappropriate compromise, and disregard of the reasonable-doubt requirement) more substantial than the risk of chilling the jury’s equity-dispensing function?